Has economic policy uncertainty slowed down the world economy?

Transcription

1 Background Paper Has economic policy uncertainty slowed down the world economy? Nicholas Bloom Stanford University

2 Has Economic Policy Uncertainty Slowed Down the World Economy? Nicholas Bloom (Stanford), 16 th January 2013 Summary Policy conflict and fiscal crisis in the United States and Europe have spurred concerns about economic policy uncertainty and its effects. Many policymakers, businessmen and the media argue that the political crisis in Washington and Brussels is leading firms and consumers to postpone hiring and spending decisions, stalling the recovery from recession. This box seeks to investigate this through answering a series of questions, drawing heavily on the academic paper Baker, Bloom and Davis (2013) and the data provided online at Measuring economic policy uncertainty Measuring economic policy uncertainty is not easy uncertainty is a subjective concept in the minds of individuals. But we can generate a proxy for policy uncertainty, using three groups of observable measures and combining them into one index of Economic Policy Uncertainty (EPU). The first component quantifies newspaper coverage of policy-related economic uncertainty. This uses computer searches of 10 major US newspapers (such as the New York Times, Wall Street Journal, USA Today, etc.) for articles that mention uncertain* 1, econom* and one of 6 policy words congress, deficit, federal reserve, legislation, regulation and White House. This is normalized by the count of all articles in each paper month, then normalized to have a standard deviation of 1 in each paper, and then added up across all 10 papers. The second component reflects the number and size of federal tax code provisions set to expire in future years. It uses the Congressional Budget Office s list of the dollar value of expiring tax code provisions, and a 50% annual discount rate (thereby focusing on nearterm expirations) to generate a tax expiration index. The third component captures disagreement among economic forecasters in the Philadelphia Federal Reserve s Survey of Professional Forecasters. This component combines the interquartile range of 1 year-ahead quarterly forecasts of Federal, State and Local Government purchases (normalized by a backward looking moving average of GDP) and the interquartile range of Consumer Price Index forecasts To create our overall measure we put weight 1/2 on the news index, and 1/6 on each of the tax-expiration, government purchases disagreement and CPI disagreement indices, 1 The * denotes any ending for example uncertain, uncertainty or uncertainties.

3 after these have all been normalized to have a standard deviation of 1. The index is then finally normalized to have a value of 100 prior to US, European and Chinese Policy Uncertainty The US policy uncertainty index is shown in Figure 1. This clearly bounces around over time, surging around major wars, elections and terrorist attacks. But US policy uncertainty has recently jumped up and remained high since As I discuss below I think this survey in policy uncertainty is unfortunately here to stay as the factors driving this look depressingly permanent. Using the news search data we can also break down the key policy terms leading to changes in uncertainty over time. We find that in the US tax, government spending and healthcare regulatory uncertainty are the key factors driving the increases since Newspaper articles seems to discuss frequently how firms and individuals have little clarity about future tax rates and government spending, and worry about the passage of the Affordable Care Act which Democrats and Republicans have been battling over. Intriguingly, monetary policy uncertainty does not appear to have risen particularly dramatically since This seems surprising given how active the Fed have with several rounds of interest rate cuts and unconventional policies like quantitative easing. But is appears that because inflation and interest rates are low and stable the mainstream US media does not perceive any increase in monetary policy uncertainty. Interestingly, the US media is also not that concerned with Europe the coverage of overseas economic policy uncertainty is very limited, with most of the rise since 2008 focused on domestic US policies. Europe s index of economic policy uncertainty is plotted in Figure 2 based on a 50% mix of news articles in the 10 major European newspapers and 50% of disagreement of forecasters over future Government expenditure and CPI (there are no major expiring tax provisions in Europe so no index of this). The data shows a similar time profile to the US with a surge post 2008, suggesting both European and US policy uncertainty have contributed to increased global policy uncertainty since Unlike US newspapers, however, European newspapers do have a global outlook and are clearly impacted by US policy uncertainty. China s index is shown in Figure 3 and shows some similarity to the US and Europe with recently very high levels. However, Chinese policy uncertainty also dances to its own tune. The November 2012 leadership handover in China appears to be about as important as the global economic slowdown for policy uncertainty. China also had large earlier spikes after events like the Oct 2001 hand-over of Hong Kong and the inflationary pressures in late In Baker, Bloom and Davis (2013) we also do some methodological checks of the index. For example, we compare our index against the frequency of the word uncertainty in the Federal Open Market Committee s (FOMC s) Beige Book, which is a 15,000 word summary of the US economy published before every FOMC rate setting meeting. We count both the frequency of the word uncertain and also had an undergraduate read the

4 Beige Book and categorize every appearance of the word uncertainty into a policy or non-policy context. Figure 4 plots the frequency of uncertainty mentions and policy context uncertainty mentions. It is clear the frequency of uncertainty arising in the Beige Book shows a very similar pattern to our overall US policy uncertainty index (correlation=0.802), with high levels post 2008 and smaller spikes after Gulf War I, II and 9/11. The impact of economic policy on the economy There are theoretically many reasons why uncertainty can damage growth. In fact one of the earliest papers in the economics literature is Bernanke (1983), written by the current Chairman of the Federal Reserve Board Ben Bernanke, pointing out that increases in uncertainty lead firms to defer investment, creating short, sharp recessions. More recently Bloom (2009) has shown large uncertainty shocks in the US can lead to sharp recessions as firms and consumers put spending plans on hold. This occurs both because uncertainty makes firms more cautious about investing and hiring, but also because it make it harder to raise finance. Banks are less willing to lend to firms in uncertain periods, squeezing the ability of companies to invest. But how big is the negative impact of policy uncertainty? To attempt to estimate the impact of policy uncertainty Baker, Bloom and Davis (2013) run vector auto regressions (VAR) estimations for US GDP growth and employment on our US index for economic policy uncertainty, plus other key economic factors like interest rates, inflation and stock-market levels. Figure 5 shows our basic result from these regressions increases in policy uncertainty forecast future drops in economic growth and employment. These magnitudes are also large for example, the figures show that following an increase in policy uncertainty of the size seen on average between 2006 and 2011, industrial production drops by 2.5% and employment by 2.4m. These results are also robust to changes in the VAR estimation for example, adding or removing different variables, changing variable ordering or timing or including controls for other measures like the Michigan consumer confidence index generates similar findings (see Figure 6). However, an obvious concern is over causality. Since policy making is forward looking and responds to the economic cycle, perhaps our VAR-based results simply reflect a tendency for policy to become more volatile and unpredictable when an economic downturn looms on the horizon. So we can also examine qualitative evidence surveys from businesses and consumers about the factors influencing their decisions, which also suggest a role for policy uncertainty. For example, the 2012 National Federation of Independent Businesses small firm survey reports that 35% of small firms complain about uncertainty of government actions as a critical problem. But what about the rest of the world - how will it be impacted by US and European policy uncertainty? There are two main channels of likely spillover. The first is through demand and FDI. The US and Europe account for more than half of all world trade and outbound FDI, and if they slow down this spills over globally through lower exports and

5 multinational investment. Second, they are also the major centers of finance and if Wall Street and European stock markets become more volatile due to higher policy uncertainty this is likely to have a global contagion effect. So unfortunately the rest of the world has been and will continue to be negatively impacted by policy uncertainty in the US and Europe. Can firms and individuals do anything to reduce the impact of policy uncertainty The implications for firms and individuals are two fold. First, be more cautious. As shown by Dixit and Pindyck (1994) when conditions are more uncertain it pays to be risk averse when making irreversible decisions. So firms should reconsider investment and hiring decisions (increase their required returns on investment by adding an uncertainty premium), while consumers should carefully evaluate expenditures particularly on durable products like cars and household goods. The reason is if US or European policy suddenly radically changes, causing the global economy to lurch, it will be costly to have made major decisions that are expensive to unwind. Second, firms and consumers should try to actively manage risks. For example, attempt to reduce exposures to the most sensitive sectors like US defense and healthcare or Greek markets. This will help to lower minimize the impact of international policy risk. What about the future - will policy uncertainty remain high? The prospects for US policy uncertainty falling in the near term do not appear bright. There are two reasons for pessimism one linked to the current political agenda, and the other to the polarization of the US political system. Policy agenda: The deal reached at the end of 2012 did little to address the long-run US structural deficit. This is both large and trending upwards due to rising healthcare costs, primarily arising from Medicare. To address this either Medicare costs have to be contained, or taxes have to be raised substantially. The Democrats have fiercely resisted extensive controls on Medicare expenses while the Republicans have fiercely resisted extensive increases in taxes. As such, the current scenario of lurching from policy crisis to policy crisis as each new debt-ceiling deadline arises appears likely to continue. This is not a particularly US problem many other OECD countries have budgetary challenges but because US expenditure on healthcare as a share of GDP at 19%, approximately double the OECD average, these healthcare cost challenges are acute in the US. Polarization: Another factor that is hampering the ability to reach a compromise in Washington is the increasing polarization of the political system, In the House of Representatives redistricting has led to an increasing share of congressional districts that are either strongly Democrat or strongly Republican. This means that the biggest threat to incumbents is not challenges from the opposition party but from other party members in their primaries. As a result politicians have little incentive to move to the middle of the political spectrum to defend against the opposition party, but instead want to move to the middle of their own party activists political spectrum. This is leading to an increasingly large gap between the policies and voting behavior of the two parties. For example, Caroll et al (2008) report that while the 90 th congress of 1967/68 showed considerable

6 overlap in the voting patterns of Democrats and Republicans the 110 th congress of 2007/2008 showed almost no overlap. Hence, getting agreement on policies in the House and even the Senate is becoming harder and harder over time as politicians polarize into two camps. In Europe prospects are unfortunately equally bleak. European policy uncertainty is high because Southern European countries need to undertake massive deregulation and reform to increase productivity growth if they are to stay within the Euro. 2 But in Southern Europe pro-market reforms are politically unpopular, making the process perilous and uncertain. Without these reforms these countries will eventually have to leave the Euro, destabilizing their economies and those of Northern Europe as well. Whether or not these reforms will occur or the countries will need to exit the Euro is going to take some years to play out as, foreboding many more years of extended policy uncertainty in Europe. So looking ahead the outlook is bleak in the US and Europe. Businesses and consumers in most OECD countries are unfortunately going to have to live with many more years of heightened policy uncertainty. Economic policy uncertainty is the new normal. Bibliography: Baker, S., Bloom, N. and Davis, S. (2013), Measuring Economic Policy Uncertainty, Stanford mimeo Baker, Scott, Nicholas Bloom, Steven J. Davis, and Xiaoxi Wang, "A Measure of Economic Policy Uncertainty for China," Work in progress, University of Chicago Baker, S., Bloom, N. and Davis, S. (2013b), What triggers stock market jumps, Stanford mimeo Bloom, Nick. (2009): The Impact of Uncertainty Shocks, Econometrica, 77, pp Carroll, Royce, Jeffrey Lewis, James Lo, Nolan McCarty, Keith Poole and Howard Rosenthal (2008) Who Is More Liberal, Senator Obama or Senator Clinton?, working paper, 18 April. Dixit, Avinash and Pindyck, Robert (1994), Investment Under Uncertainty, Princeton University Press, Princeton, New Jersey. 2 The reason is because without these deregulations Southern European productivity growth is likely to continue to be a couple of percent lower than Germany s, requiring Southern European wage growth to be similarly lower than Germany s. But since German inflation is low and Southern Europe s is typically higher, this generates a mismatch with the fixed exchange rates of the Euro system.

7

8

9

10

11

12